Posts filed under “Data Analysis”
The WSJ’s Greg Ip is rapidly becoming a proverbial two-handed economist. Over the past week, he has written not one but two articles on the ambiguous nature of the macro-data that has been coming forward over the past few quarters.
The first article was last week’s Fitting Growth Data Into the Unemployment Puzzle and is timely, with prelim GDP data out tomorrow. This is fascinating issue I have been watching for many quarters now: When two (allegedly) reliable historical data points are in direct conflict, which one is correct? What should get more emphasis, and which should gets less?
"When the Commerce Department publishes first-quarter growth
estimates Friday, it will accentuate a puzzle with implications for interest
rates, growth and U.S. living standards: With the economy growing so slowly, why
is unemployment falling?
Explanations range from measurement problems to the peculiar
behavior of construction jobs to the possibility that productivity growth is
Wall Street estimates the economy grew 1.9% at an annual rate in
the first quarter, the slowest in more than a year. That would bring growth for
the past four quarters down to a sluggish 2.2%, below most economists’ estimates
of the economy’s potential growth rate, which hovers around 3%. Potential is the
rate at which the economy can grow over the long term, given growth in the work
force and worker productivity.
A year of below-potential growth ordinarily would push up
unemployment, because the economy isn’t growing fast enough to employ all the
new entrants to the work force. Instead, the unemployment rate has fallen to
4.4% in March from 4.7% a year earlier.
The answer to this puzzle has huge implications for inflation and
interest rates. For the Federal Reserve, the most troubling explanation is that
potential growth is closer to 2% than 3%. That may be because the work force is
growing more slowly for demographic reasons, or because productivity growth has
slowed, perhaps because companies are investing less in labor-saving
There are several potential answers to this puzzle:
1) Growth is not slowing, GDP measures ares all wrong.
2) Unemployment is not falling, BLS measures are incorrect.
3) The relationship between GDP and Unemployment have decoupled.
We know that growth overseas has been nothing short of robust — and perhaps globalization is seeing some spillover into the US economy in ways GDP doesn’t measure. That’s one possibility.
While the Unemployment Rate has notched down, new unemployment claims have moved higher; At the same time, the exhaustion rate — the measure of those people who have used up all their unemployment insurance benefits but have not found a job — is also higher. Another explanation is that the labor participation rate is falling as some unemployed leave the labor market entirely.
Regardless of the explanation, these two data series seem to be in conflict. Traditionally, there should be a greater correlation between the two, and this implies something is off: Either Growth is better than reported, or Employment is worse than reported.
On a related (two-handed) note, Ip follows up this morning with another interesting quandry: Why Market Strength And Economic Growth Don’t Always Line Up. He pits Merrill Lynch’s David Rosenberg against ISI’s Ed Hyman — thats a heavy wight battle for ya — on what this means to the market:
"Economists differ. David Rosenberg, chief North
American economist at Merrill Lynch & Co., thinks the market is
misleadingly optimistic. He says there is a "disconnect between how the
economy is doing and the way the equity market is doing." The U.S.
economy has just completed four quarters of annualized growth below 3%,
which, he says, has never happened in 60 years without being followed
by recession. While he doesn’t forecast one, he puts the risks higher
than generally realized.
But Ed Hyman, chief economist at ISI Group, a New York
investment dealer, says the market has it right. Just as in 1985 and
1995, the Federal Reserve has raised interest rates enough to slow the
economy and bring inflation under control. That reassures investors
that even higher rates won’t be needed later that could tip the economy
The full piece is worth a read . . .
UPDATE: April 26, 2007 10:06am
A perfect example of both the Employment/GDP issue as well as the Global/Domestic divergment is Monster Worldwide — they reported their quarterly profit fell — their sales rose 28%, in large part due to a 64% jump in international careers business. Overseas sales is now over a third of their revenues . . .
Fitting Growth Data Into the Unemployment Puzzle (free WSJ)
WSJ, April 21, 2007; Page A2
Why Market Strength And Economic Growth Don’t Always Line Up
WSJ April 26, 2007; Page C1